Obesity is perhaps the greatest health threat facing the US, and policy-based prevention approaches must be part of the solution. In preschool-aged children, regular exercise decreases obesity risk, improves motor skills, and enhances cardiovascular function. With nearly three quarters of young children in some form of non- parental child care and 56% in center-based child care, the child care setting may be a key to reversing disturbing trends in physical activity and obesity. Massachusetts (MA) recently implemented a new policy requiring 60 minutes of daily physical activity for all children in child care. A narrow window of opportunity now exists to evaluate this new policy, which will affect over 2,000 child care centers caring for nearly 40,000 children in MA. We intend to leverage this natural experiment to assess the effectiveness of the new policy by measuring physical activity in children attending child care center before and after the policy takes effect, both in MA and Rhode Island (RI), a comparison state. Child care is not regulated by the federal government-each individual state is responsible for regulating its child care centers. Though 2 states, Delaware and Alaska, have recently enacted new policies requiring 60 minutes of daily physical activity for all children in child care, they have not been evaluated for effectiveness. The proposed study provides a unique opportunity to evaluate this physical activity policy. The primary aim of the study is to determine the extent to which the new policy in MA, compared with RI, increases physical activity in 3- to 5-year old children in child care. We will use a quasi-experimental two-group pre-test/post-test design to assess physical activity in a sample of children attending child care centers before and after the policy takes effect. This evaluation is already underway. We collected 3 rounds of pre-policy physical activity data over the course of 1.5 years (fall 2008, spring 2009, and fall 2009). We focused our data collection efforts on Boston, MA and Providence, RI and measured physical activity in a sample of children attending 40 child care centers (20 per city). We collected both observation (n=360 children) and accelerometer (n=41 children) data to assess the physical activity levels of children, focusing on total minutes children engage in moderate to vigorous physical activity. In this proposal, we seek funding to support post- policy physical activity data collection to complete our evaluation of this natural experiment.